Wafer Inspection System for Probe Mark Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual inspection of probe marks on wafers is time-consuming and prone to human error, leading to yield loss due to inconsistent standards and the inability to efficiently identify unqualified devices under test (DUTs).
Innovation Solution
A wafer inspection system that includes a memory unit, image-uploading unit, and processing unit to identify candidate regions, generate confidence scores for probe marks, select the highest scoring regions, and eliminate overlapping regions, utilizing artificial intelligence to improve accuracy and reduce human error.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual inspection is used to inspect probe marks, then flexibility and adaptability are maintained, but inspection time increases and human error occurs
Solution Approach 1:
The patent replaces the manual mechanical inspection system with an automated image processing system that captures wafer images, identifies probe marks through algorithmic analysis, and determines qualification status automatically. This substitution eliminates human factors causing time delays and errors while maintaining consistent inspection criteria.
Solution Approach 2:
The inspection system performs self-evaluation by automatically analyzing captured images, identifying probe marks, and determining whether DUTs meet specifications without requiring human intervention. The system serves itself by generating inspection results through programmed algorithms that consistently apply qualification criteria.
2Reliability
If manual inspection is performed by multiple engineers, then comprehensive coverage is achieved, but inconsistency in standards and human error increase
Solution Approach 1:
The patent implements a universal inspection system that handles all probe mark inspections using the same automated image processing algorithms. This single system replaces multiple engineers, ensuring that identical qualification criteria are applied consistently across all inspections without variation in human judgment or interpretation.
Solution Approach 2:
The system transforms the inspection process from subjective human judgment to objective parameter-based analysis by capturing images and applying algorithmic criteria for identifying probe marks and determining qualification. This parameter-based approach eliminates inconsistencies in human standards while maintaining clear, defined inspection criteria.
3Productivity
If probe marks are left on DUT during testing, then electrical characteristics can be tested, but bonding pad quality issues and packaging failure occur
Solution Approach 1:
The patent performs preliminary inspection of probe marks immediately after testing while the wafer is still in position. By capturing images and analyzing probe mark conditions before the DUTs are moved to subsequent processing steps, the system identifies and flags affected DUTs early, preventing further damage during handling and enabling timely corrective actions.
Solution Approach 2:
The inspection system provides feedback by automatically generating qualification results based on probe mark analysis. This feedback mechanism identifies DUTs with bonding pad quality issues or packaging failure risks, allowing manufacturers to sort and handle affected devices differently, thereby mitigating the harmful effects of probe marks before they cause additional damage.
Data Source
AI summary
A wafer inspection system is provided. The wafer inspection system includes a memory unit configured to store an image of a device under test (DUT) on a wafer, an image-uploading unit configured to upload the image to a processing unit, and a processing unit. The processing unit is configured to identify a plurality of candidate regions on the image; generate a confidence score for each of the plurality of candidate regions, wherein the confidence score indicates a probability of a candidate region including a probe mark; select a first candidate region having the highest confidence score as a selected region; determine whether a second candidate region in the plurality of candidate regions includes the same probe mark as the first candidate region; and eliminate the second candidate region if the second candidate region includes the same probe mark as the first candidate region.